The goal of this project is to relate physiological and psychoacoustical measures of responses to temporal envelope fluctuations. Behaviorally-relevant sounds, such as speech, contain many amplitude-modulated features, but the neural mechanisms that underlie their perception remain unclear. This proposed research will critically test both existing and novel encoding hypotheses using human psychophysics, awake rabbit physiology, and computational modeling of auditory processing. By using the same type of stimulus in each of the three experimental paradigms, findings from one approach can be directly integrated into tests of the others. Audio-frequency stimuli that have been useful in determining the perceptual relevance of proposed decision variables will be translated into the modulation-frequency domain. One set of experiments will study the representation of sinusoidal modulation in the presence of an additional masking noise modulation. Masked-modulation detection thresholds have been measured psychophysically, and responses of inferior colliculus neurons will be obtained in matching stimulus conditions. Comparisons of competing model predictions will lend insight into cues used by the real system